Dilated DenseNets for Relational Reasoning
Antreas Antoniou, Agnieszka S{\l}owik, Elliot J. Crowley, Amos Storkey
TL;DR
This paper demonstrates that a DenseNet with dilated convolutions can perform relational reasoning effectively on the Sort-of-CLEVR dataset, eliminating the need for expensive relational modules.
Contribution
The authors introduce a DenseNet architecture with dilated convolutions that achieves relational reasoning without additional relational modules.
Findings
Outperforms models with explicit relational modules on Sort-of-CLEVR
Reduces computational complexity in relational reasoning tasks
Shows effectiveness of dilated DenseNets for relational reasoning
Abstract
Despite their impressive performance in many tasks, deep neural networks often struggle at relational reasoning. This has recently been remedied with the introduction of a plug-in relational module that considers relations between pairs of objects. Unfortunately, this is combinatorially expensive. In this extended abstract, we show that a DenseNet incorporating dilated convolutions excels at relational reasoning on the Sort-of-CLEVR dataset, allowing us to forgo this relational module and its associated expense.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
Taxonomy
TopicsTopic Modeling · Multimodal Machine Learning Applications · Natural Language Processing Techniques
Methods*Communicated@Fast*How Do I Communicate to Expedia? · Batch Normalization · Convolution · Average Pooling · Concatenated Skip Connection · Global Average Pooling · Dense Block · Kaiming Initialization · 1x1 Convolution · Dropout